Folding Forum

Here are some WUs that four clients completed in the same day. The project has a timeout of two days. This looks like something malfunctioning and reassigning the WU way before it times out. It's a waste of computing resources to do the same work multiple times. https://apps.foldingathome.org/wu?p=1...

Server vav20.fah.temple.edu (129.32.209.203) needs fixing. 11:58:30:WU01:FS00:Sending unit results: id:01 state:SEND error:NO_ERROR project:12418 run:126 clone:2 gen:265 core:0xa8 unit:0x09010000020000007e00000082300000 11:58:30:WU01:FS00:Uploading 9.55MiB to 129.32.209.203 11:58:30:WU01:FS00:Connec...

It's still undeniable that people who run potato farms with slow hardware are actually slowing down the scientists who need the results to come back quickly before new WUs can be generated. This is why the quick return bonus exists. People are taking 2-4 days to process when someone else with faster...

Some of your WUs are being returned late and end up being sent out to be run by another user. Three examples just from recent days: https://apps.foldingathome.org/wu?p=18494&r=10&c=49&g=1#project=18494&run=10&clone=49&gen=1 https://apps.foldingathome.org/wu?p=18494&r=16&a...

Looks like the ocl-icd-opencl-dev package needs to be installed for the OpenCL libary. For CUDA support, the NVIDIA drivers would need to be installed. CUDA is more efficient than OpenCL. Hopefully your electricity costs are low. A card this old isn't that efficient compared to modern one. A 2060 wo...

Others have stated that RPI4 was too slow to complete Folding tasks long ago. It's time to retire or repurpose this hardware.

viewtopic.php?p=365160&hilit=Raspberry+Pi#p365160
https://www.reddit.com/r/Folding/commen ... is_remote/

It falls back to using OpenCL if the CUDA libraries can't be loaded. CUDA has significantly better performance than OpenCL, so it's better to use that unless you have some kind of philosophical objection to something proprietary.

FAHControl can be used to control remote clients if port 36330 is opened. Not sure if remote web client works.

Client 7.6.21. The client version shouldn't matter since the downloaded core binaries are what process the work units.

On Fedora 40, 4070 Super does 13-14M PPD most days, rare days of 15-16M. And 4070 Ti does 14-15M PPD most days, 13M or 16M rarely. Using recent kernels from Fedora repo and latest NVIDIA drivers from rpmfusion repo.

Just got similar crashes with these:

Project: 19230 (Run 783, Clone 20, Gen 0)
Project: 19230 (Run 1023, Clone 20, Gen 0)

Here's a comparison to newer hardware and the typical power:

E5-2690 v4: 400-450K PPD @200W
E5-2697a v4: 500-550K PPD @200W
Ryzen 9 7950X: 1-1.4M PPD @200W (CPU temp limited to 85C in BIOS)

RTX 4060 Ti: 6-7M PPD @110-150W
RTX 4070 Super: 13-14M PPD @170-200W
RTX 4070 Ti: 14-16M PPD @180-200W

This WU segfaults and restarts endlessly. The client is unable to detect that it's crashing multiple times and then halt it. 23:10:58:WU02:FS00:0xa8:Project: 19230 (Run 6802, Clone 20, Gen 0) 23:10:58:WU02:FS00:0xa8:Unit: 0x00000000000000000000000000000000 23:10:58:WU02:FS00:0xa8:Reading tar file co...

Well I care about doing the most scientific research possible for the money. That requires at least semi-modern, used hardware like Ryzen or Xeon CPUs with a lot of cores. Quad or dual core is just a waste of power for distributed computing. I've slowly gone from 4 to 6, 8, 14 and finally 16-core CP...

It's more than just speed that's a drawback. Old hardware wastes power and costs money for the utility bill while doing very little work in the process.